Video Editor Online

• Outcome: Review
• Latest audit: May 29, 2026 at 9:32 PM UTC
• Version:
  1.0.3

Overview

This is a real cloud video-editing skill, but it automatically creates remote authenticated sessions and may send broad prompts or media to a third-party backend with limited user-facing disclosure.

Review before installing. Use this only if you are comfortable with Nemovideo receiving editing prompts, uploaded files or URLs, project state, and render outputs. Prefer a dedicated or low-privilege NEMO_TOKEN, avoid confidential or regulated footage unless you trust the provider’s retention and privacy practices, and confirm that any URL or file upload is intentional.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Context-Inappropriate Capability

Medium

• Confidence: 91% confidence
• Finding: The skill instructs the agent to automatically consume a bearer token from the environment and, if absent, to mint a new anonymous token and create a remote session before handling requests. That causes credential use and outbound authentication to occur implicitly, which is risky because the skill is framed as an editing copilot rather than explicitly requiring informed consent for account/session creation and token-backed remote operations.

Context-Inappropriate Capability

Medium

• Confidence: 84% confidence
• Finding: Allowing upload/import from arbitrary remote URLs expands the skill from local editing assistance into generalized remote content retrieval. This can be abused to cause the backend to fetch attacker-controlled or sensitive internal URLs, or to ingest untrusted content without clear user understanding of the privacy and security implications.

Vague Triggers

Medium

• Confidence: 88% confidence
• Finding: The routing table includes a catch-all fallback of 'Everything else' to the SSE pipeline, which can cause unrelated user inputs to be sent to the remote backend. In a skill that transmits data off-platform, overly broad activation increases the chance of accidental data disclosure and unintended external actions.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The skill directs the agent to generate a client identifier, obtain an anonymous token, and create a session without a user-facing warning or consent step. This is dangerous because it silently transmits identifiers and credentials to a third-party service, undermining user expectations around privacy and external account/session creation.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill describes cloud upload, rendering, staging, and download of user media but provides no prominent privacy or data-handling disclosure. Because videos commonly contain sensitive personal, corporate, or copyrighted material, silently sending them to a remote backend materially increases confidentiality and compliance risk.

VirusTotal

66/66 vendors flagged this skill as clean.